﻿1
00:00:00,433 --> 00:00:01,533
Hello and welcome back.

2
00:00:01,533 --> 00:00:02,900
Let's have a quick look at.

3
00:00:02,900 --> 00:00:04,666
Multiple linear regression.

4
00:00:04,666 --> 00:00:07,066
So here is the equation
for multiple linear regression.

5
00:00:07,066 --> 00:00:10,533
As you can see is quite similar
to the simple linear regression equation.

6
00:00:10,900 --> 00:00:12,433
Here we've got the dependent variable.

7
00:00:12,433 --> 00:00:14,500
Then we've got the y
intercept of the constant.

8
00:00:14,500 --> 00:00:17,500
Then we've got a slope coefficient
an independent variable.

9
00:00:17,600 --> 00:00:19,600
And then we've got more pairs of those.

10
00:00:19,600 --> 00:00:22,700
Then we've got another slope coefficient
an independent variable and so on.

11
00:00:22,700 --> 00:00:25,700
Add another slope coefficient
an independent variable.

12
00:00:25,700 --> 00:00:30,500
So basically we have as many slope
coefficients as we have independent

13
00:00:30,500 --> 00:00:31,833
variables.

14
00:00:31,833 --> 00:00:35,766
Now going back to our example
of producing potatoes on a farm,

15
00:00:36,066 --> 00:00:40,233
we might want to predict how many potatoes
we are yielding

16
00:00:40,233 --> 00:00:45,433
depending on how many kilograms
of nitrogen fertilizer we're using.

17
00:00:45,633 --> 00:00:51,766
What the average temperature, in
the season is and what the millimeter is,

18
00:00:51,766 --> 00:00:55,700
how many millimeters of rainfall,
we've seen in the season.

19
00:00:56,200 --> 00:01:00,366
So the potatoes in this case,
the equation for potatoes would be,

20
00:01:01,133 --> 00:01:04,300
for example, eight tonnes would be a
y intercept.

21
00:01:04,666 --> 00:01:08,633
then three tonnes per kilogram
would be the amount of fertilizer,

22
00:01:09,233 --> 00:01:12,366
the, coefficient for fertilizer used.

23
00:01:12,633 --> 00:01:14,933
Then for average temperature in degrees
Celsius,

24
00:01:14,933 --> 00:01:18,900
we might have a -0.54 tonnes
per degrees Celsius.

25
00:01:18,900 --> 00:01:19,533
And what that means

26
00:01:19,533 --> 00:01:23,800
is that the higher the temperature,
the less potatoes will be yielded.

27
00:01:24,200 --> 00:01:28,266
and then for the rainfall might be 0.04
tonnes per millimeter.

28
00:01:28,866 --> 00:01:33,500
Now, to finish off, what I wanted to say
is that, this we won't be using

29
00:01:33,500 --> 00:01:36,833
this example in all practical terms
will be a different example.

30
00:01:36,866 --> 00:01:39,100
It'll be fun
and interesting example as well.

31
00:01:39,100 --> 00:01:42,100
But if you wanted to learn more
about potatoes

32
00:01:42,100 --> 00:01:45,133
and specifically the applications
of multiple linear regression,

33
00:01:45,366 --> 00:01:48,766
here's a research paper
that you might be interested in.

34
00:01:49,333 --> 00:01:51,766
it's called the Application
of Multiple Linear regression

35
00:01:51,766 --> 00:01:53,666
and artificial Neural network models

36
00:01:53,666 --> 00:01:57,633
for yield prediction of very early potato
cultivators before harvest.

37
00:01:57,966 --> 00:02:02,766
So, this paper combines a multiple linear
regression and neural networks.

38
00:02:02,766 --> 00:02:05,733
It looks at two different ways
of modeling them.

39
00:02:05,733 --> 00:02:08,633
of course you can just focus on the
multiple linear regression if you like.

40
00:02:08,633 --> 00:02:12,900
It's quite an interesting read and
it can give you some additional, deeper

41
00:02:13,633 --> 00:02:18,033
knowledge into the world of harvesting
potatoes and multiple linear regression.

42
00:02:18,033 --> 00:02:19,000
There.

43
00:02:19,000 --> 00:02:21,900
Other than that, enjoy
the practical tutorials and for lunch,

44
00:02:21,900 --> 00:02:23,600
and I'll see you back here next time.

45
00:02:23,600 --> 00:02:25,466
Until then, enjoy machine learning.
